{
  "best_metric": 0.21390262246131897,
  "best_model_checkpoint": "finetuned-indian-food/checkpoint-1300",
  "epoch": 4.0,
  "global_step": 1332,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "learning_rate": 0.0001984984984984985,
      "loss": 2.8538,
      "step": 10
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00019699699699699701,
      "loss": 2.5964,
      "step": 20
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0001954954954954955,
      "loss": 2.2524,
      "step": 30
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00019399399399399402,
      "loss": 1.9596,
      "step": 40
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0001924924924924925,
      "loss": 1.726,
      "step": 50
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.000190990990990991,
      "loss": 1.6152,
      "step": 60
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0001894894894894895,
      "loss": 1.3396,
      "step": 70
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.000187987987987988,
      "loss": 1.2854,
      "step": 80
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001864864864864865,
      "loss": 1.107,
      "step": 90
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.000184984984984985,
      "loss": 1.0846,
      "step": 100
    },
    {
      "epoch": 0.3,
      "eval_accuracy": 0.8554729011689692,
      "eval_loss": 0.9561436772346497,
      "eval_runtime": 39.5068,
      "eval_samples_per_second": 23.819,
      "eval_steps_per_second": 2.987,
      "step": 100
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0001834834834834835,
      "loss": 0.9812,
      "step": 110
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.000181981981981982,
      "loss": 0.8571,
      "step": 120
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0001804804804804805,
      "loss": 0.916,
      "step": 130
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00017897897897897898,
      "loss": 0.9905,
      "step": 140
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001774774774774775,
      "loss": 0.7381,
      "step": 150
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.000175975975975976,
      "loss": 0.801,
      "step": 160
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0001744744744744745,
      "loss": 0.6606,
      "step": 170
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.000172972972972973,
      "loss": 0.7538,
      "step": 180
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00017147147147147148,
      "loss": 0.6495,
      "step": 190
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00016996996996997,
      "loss": 0.7894,
      "step": 200
    },
    {
      "epoch": 0.6,
      "eval_accuracy": 0.8926673751328374,
      "eval_loss": 0.587147057056427,
      "eval_runtime": 38.4761,
      "eval_samples_per_second": 24.457,
      "eval_steps_per_second": 3.067,
      "step": 200
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00016846846846846846,
      "loss": 0.7121,
      "step": 210
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00016696696696696697,
      "loss": 0.6888,
      "step": 220
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00016546546546546546,
      "loss": 0.5768,
      "step": 230
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00016396396396396395,
      "loss": 0.6568,
      "step": 240
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00016261261261261262,
      "loss": 0.601,
      "step": 250
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0001611111111111111,
      "loss": 0.5431,
      "step": 260
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00015960960960960963,
      "loss": 0.6634,
      "step": 270
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.00015810810810810812,
      "loss": 0.8153,
      "step": 280
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00015660660660660663,
      "loss": 0.9727,
      "step": 290
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.00015510510510510512,
      "loss": 0.6233,
      "step": 300
    },
    {
      "epoch": 0.9,
      "eval_accuracy": 0.9107332624867163,
      "eval_loss": 0.4447122812271118,
      "eval_runtime": 38.3227,
      "eval_samples_per_second": 24.555,
      "eval_steps_per_second": 3.079,
      "step": 300
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0001536036036036036,
      "loss": 0.578,
      "step": 310
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.00015210210210210212,
      "loss": 0.6366,
      "step": 320
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.0001506006006006006,
      "loss": 0.8622,
      "step": 330
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00014909909909909913,
      "loss": 0.5957,
      "step": 340
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00014759759759759762,
      "loss": 0.4458,
      "step": 350
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0001460960960960961,
      "loss": 0.383,
      "step": 360
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.00014459459459459462,
      "loss": 0.4259,
      "step": 370
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.00014309309309309308,
      "loss": 0.4246,
      "step": 380
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0001415915915915916,
      "loss": 0.3343,
      "step": 390
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0001400900900900901,
      "loss": 0.3619,
      "step": 400
    },
    {
      "epoch": 1.2,
      "eval_accuracy": 0.8937300743889479,
      "eval_loss": 0.4354824125766754,
      "eval_runtime": 37.6727,
      "eval_samples_per_second": 24.978,
      "eval_steps_per_second": 3.132,
      "step": 400
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.00013858858858858858,
      "loss": 0.5238,
      "step": 410
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0001370870870870871,
      "loss": 0.5456,
      "step": 420
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.00013558558558558558,
      "loss": 0.5347,
      "step": 430
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0001340840840840841,
      "loss": 0.3739,
      "step": 440
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.00013258258258258258,
      "loss": 0.392,
      "step": 450
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.00013108108108108107,
      "loss": 0.5306,
      "step": 460
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0001295795795795796,
      "loss": 0.4171,
      "step": 470
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.00012807807807807808,
      "loss": 0.3715,
      "step": 480
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.00012657657657657657,
      "loss": 0.5192,
      "step": 490
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.00012507507507507508,
      "loss": 0.34,
      "step": 500
    },
    {
      "epoch": 1.5,
      "eval_accuracy": 0.9117959617428267,
      "eval_loss": 0.3712367117404938,
      "eval_runtime": 37.8371,
      "eval_samples_per_second": 24.87,
      "eval_steps_per_second": 3.119,
      "step": 500
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.00012357357357357357,
      "loss": 0.5209,
      "step": 510
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.00012207207207207209,
      "loss": 0.4896,
      "step": 520
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.00012057057057057057,
      "loss": 0.5666,
      "step": 530
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.00011906906906906908,
      "loss": 0.3288,
      "step": 540
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.00011756756756756758,
      "loss": 0.4315,
      "step": 550
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.00011606606606606607,
      "loss": 0.4185,
      "step": 560
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.00011456456456456457,
      "loss": 0.2794,
      "step": 570
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.00011306306306306307,
      "loss": 0.3125,
      "step": 580
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.00011156156156156157,
      "loss": 0.3229,
      "step": 590
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.00011006006006006006,
      "loss": 0.3413,
      "step": 600
    },
    {
      "epoch": 1.8,
      "eval_accuracy": 0.8916046758767269,
      "eval_loss": 0.40877240896224976,
      "eval_runtime": 39.2551,
      "eval_samples_per_second": 23.971,
      "eval_steps_per_second": 3.006,
      "step": 600
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.00010855855855855856,
      "loss": 0.3967,
      "step": 610
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.00010705705705705707,
      "loss": 0.32,
      "step": 620
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.00010555555555555557,
      "loss": 0.3256,
      "step": 630
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.0001042042042042042,
      "loss": 0.4024,
      "step": 640
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.0001027027027027027,
      "loss": 0.4209,
      "step": 650
    },
    {
      "epoch": 1.98,
      "learning_rate": 0.0001012012012012012,
      "loss": 0.3888,
      "step": 660
    },
    {
      "epoch": 2.01,
      "learning_rate": 9.96996996996997e-05,
      "loss": 0.3793,
      "step": 670
    },
    {
      "epoch": 2.04,
      "learning_rate": 9.81981981981982e-05,
      "loss": 0.3094,
      "step": 680
    },
    {
      "epoch": 2.07,
      "learning_rate": 9.66966966966967e-05,
      "loss": 0.3613,
      "step": 690
    },
    {
      "epoch": 2.1,
      "learning_rate": 9.51951951951952e-05,
      "loss": 0.3619,
      "step": 700
    },
    {
      "epoch": 2.1,
      "eval_accuracy": 0.9043570669500531,
      "eval_loss": 0.37407416105270386,
      "eval_runtime": 39.0938,
      "eval_samples_per_second": 24.07,
      "eval_steps_per_second": 3.018,
      "step": 700
    },
    {
      "epoch": 2.13,
      "learning_rate": 9.36936936936937e-05,
      "loss": 0.3051,
      "step": 710
    },
    {
      "epoch": 2.16,
      "learning_rate": 9.219219219219219e-05,
      "loss": 0.4062,
      "step": 720
    },
    {
      "epoch": 2.19,
      "learning_rate": 9.069069069069069e-05,
      "loss": 0.3811,
      "step": 730
    },
    {
      "epoch": 2.22,
      "learning_rate": 8.918918918918919e-05,
      "loss": 0.3118,
      "step": 740
    },
    {
      "epoch": 2.25,
      "learning_rate": 8.76876876876877e-05,
      "loss": 0.3014,
      "step": 750
    },
    {
      "epoch": 2.28,
      "learning_rate": 8.61861861861862e-05,
      "loss": 0.2404,
      "step": 760
    },
    {
      "epoch": 2.31,
      "learning_rate": 8.468468468468469e-05,
      "loss": 0.3802,
      "step": 770
    },
    {
      "epoch": 2.34,
      "learning_rate": 8.318318318318319e-05,
      "loss": 0.3222,
      "step": 780
    },
    {
      "epoch": 2.37,
      "learning_rate": 8.168168168168169e-05,
      "loss": 0.2459,
      "step": 790
    },
    {
      "epoch": 2.4,
      "learning_rate": 8.018018018018019e-05,
      "loss": 0.2135,
      "step": 800
    },
    {
      "epoch": 2.4,
      "eval_accuracy": 0.9160467587672688,
      "eval_loss": 0.328594833612442,
      "eval_runtime": 38.4594,
      "eval_samples_per_second": 24.467,
      "eval_steps_per_second": 3.068,
      "step": 800
    },
    {
      "epoch": 2.43,
      "learning_rate": 7.867867867867868e-05,
      "loss": 0.2152,
      "step": 810
    },
    {
      "epoch": 2.46,
      "learning_rate": 7.717717717717718e-05,
      "loss": 0.2621,
      "step": 820
    },
    {
      "epoch": 2.49,
      "learning_rate": 7.567567567567568e-05,
      "loss": 0.2118,
      "step": 830
    },
    {
      "epoch": 2.52,
      "learning_rate": 7.417417417417419e-05,
      "loss": 0.2826,
      "step": 840
    },
    {
      "epoch": 2.55,
      "learning_rate": 7.267267267267268e-05,
      "loss": 0.4047,
      "step": 850
    },
    {
      "epoch": 2.58,
      "learning_rate": 7.117117117117116e-05,
      "loss": 0.4956,
      "step": 860
    },
    {
      "epoch": 2.61,
      "learning_rate": 6.966966966966967e-05,
      "loss": 0.2431,
      "step": 870
    },
    {
      "epoch": 2.64,
      "learning_rate": 6.816816816816817e-05,
      "loss": 0.2189,
      "step": 880
    },
    {
      "epoch": 2.67,
      "learning_rate": 6.666666666666667e-05,
      "loss": 0.2455,
      "step": 890
    },
    {
      "epoch": 2.7,
      "learning_rate": 6.516516516516516e-05,
      "loss": 0.2166,
      "step": 900
    },
    {
      "epoch": 2.7,
      "eval_accuracy": 0.9415515409139213,
      "eval_loss": 0.2758049964904785,
      "eval_runtime": 38.181,
      "eval_samples_per_second": 24.646,
      "eval_steps_per_second": 3.091,
      "step": 900
    },
    {
      "epoch": 2.73,
      "learning_rate": 6.366366366366366e-05,
      "loss": 0.2648,
      "step": 910
    },
    {
      "epoch": 2.76,
      "learning_rate": 6.216216216216216e-05,
      "loss": 0.2233,
      "step": 920
    },
    {
      "epoch": 2.79,
      "learning_rate": 6.0660660660660665e-05,
      "loss": 0.2876,
      "step": 930
    },
    {
      "epoch": 2.82,
      "learning_rate": 5.915915915915916e-05,
      "loss": 0.3385,
      "step": 940
    },
    {
      "epoch": 2.85,
      "learning_rate": 5.765765765765766e-05,
      "loss": 0.2823,
      "step": 950
    },
    {
      "epoch": 2.88,
      "learning_rate": 5.615615615615616e-05,
      "loss": 0.135,
      "step": 960
    },
    {
      "epoch": 2.91,
      "learning_rate": 5.465465465465466e-05,
      "loss": 0.289,
      "step": 970
    },
    {
      "epoch": 2.94,
      "learning_rate": 5.3153153153153155e-05,
      "loss": 0.2228,
      "step": 980
    },
    {
      "epoch": 2.97,
      "learning_rate": 5.165165165165166e-05,
      "loss": 0.3221,
      "step": 990
    },
    {
      "epoch": 3.0,
      "learning_rate": 5.015015015015015e-05,
      "loss": 0.1557,
      "step": 1000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9330499468650372,
      "eval_loss": 0.2679324150085449,
      "eval_runtime": 37.7775,
      "eval_samples_per_second": 24.909,
      "eval_steps_per_second": 3.124,
      "step": 1000
    },
    {
      "epoch": 3.03,
      "learning_rate": 4.8648648648648654e-05,
      "loss": 0.1808,
      "step": 1010
    },
    {
      "epoch": 3.06,
      "learning_rate": 4.714714714714715e-05,
      "loss": 0.2742,
      "step": 1020
    },
    {
      "epoch": 3.09,
      "learning_rate": 4.5645645645645645e-05,
      "loss": 0.147,
      "step": 1030
    },
    {
      "epoch": 3.12,
      "learning_rate": 4.414414414414415e-05,
      "loss": 0.248,
      "step": 1040
    },
    {
      "epoch": 3.15,
      "learning_rate": 4.264264264264264e-05,
      "loss": 0.1214,
      "step": 1050
    },
    {
      "epoch": 3.18,
      "learning_rate": 4.1141141141141144e-05,
      "loss": 0.352,
      "step": 1060
    },
    {
      "epoch": 3.21,
      "learning_rate": 3.963963963963964e-05,
      "loss": 0.2441,
      "step": 1070
    },
    {
      "epoch": 3.24,
      "learning_rate": 3.813813813813814e-05,
      "loss": 0.2205,
      "step": 1080
    },
    {
      "epoch": 3.27,
      "learning_rate": 3.663663663663664e-05,
      "loss": 0.1925,
      "step": 1090
    },
    {
      "epoch": 3.3,
      "learning_rate": 3.513513513513514e-05,
      "loss": 0.1115,
      "step": 1100
    },
    {
      "epoch": 3.3,
      "eval_accuracy": 0.9362380446333688,
      "eval_loss": 0.2528650462627411,
      "eval_runtime": 38.8623,
      "eval_samples_per_second": 24.214,
      "eval_steps_per_second": 3.036,
      "step": 1100
    },
    {
      "epoch": 3.33,
      "learning_rate": 3.3633633633633635e-05,
      "loss": 0.2245,
      "step": 1110
    },
    {
      "epoch": 3.36,
      "learning_rate": 3.213213213213213e-05,
      "loss": 0.2538,
      "step": 1120
    },
    {
      "epoch": 3.39,
      "learning_rate": 3.063063063063063e-05,
      "loss": 0.1725,
      "step": 1130
    },
    {
      "epoch": 3.42,
      "learning_rate": 2.912912912912913e-05,
      "loss": 0.3009,
      "step": 1140
    },
    {
      "epoch": 3.45,
      "learning_rate": 2.762762762762763e-05,
      "loss": 0.3042,
      "step": 1150
    },
    {
      "epoch": 3.48,
      "learning_rate": 2.6126126126126128e-05,
      "loss": 0.1748,
      "step": 1160
    },
    {
      "epoch": 3.51,
      "learning_rate": 2.4624624624624627e-05,
      "loss": 0.1229,
      "step": 1170
    },
    {
      "epoch": 3.54,
      "learning_rate": 2.3123123123123125e-05,
      "loss": 0.2376,
      "step": 1180
    },
    {
      "epoch": 3.57,
      "learning_rate": 2.1621621621621624e-05,
      "loss": 0.2025,
      "step": 1190
    },
    {
      "epoch": 3.6,
      "learning_rate": 2.012012012012012e-05,
      "loss": 0.1571,
      "step": 1200
    },
    {
      "epoch": 3.6,
      "eval_accuracy": 0.946865037194474,
      "eval_loss": 0.23596596717834473,
      "eval_runtime": 38.9005,
      "eval_samples_per_second": 24.19,
      "eval_steps_per_second": 3.033,
      "step": 1200
    },
    {
      "epoch": 3.63,
      "learning_rate": 1.8618618618618618e-05,
      "loss": 0.0491,
      "step": 1210
    },
    {
      "epoch": 3.66,
      "learning_rate": 1.7117117117117117e-05,
      "loss": 0.1921,
      "step": 1220
    },
    {
      "epoch": 3.69,
      "learning_rate": 1.5615615615615616e-05,
      "loss": 0.1733,
      "step": 1230
    },
    {
      "epoch": 3.72,
      "learning_rate": 1.4114114114114116e-05,
      "loss": 0.1822,
      "step": 1240
    },
    {
      "epoch": 3.75,
      "learning_rate": 1.2612612612612611e-05,
      "loss": 0.2661,
      "step": 1250
    },
    {
      "epoch": 3.78,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.1566,
      "step": 1260
    },
    {
      "epoch": 3.81,
      "learning_rate": 9.60960960960961e-06,
      "loss": 0.1764,
      "step": 1270
    },
    {
      "epoch": 3.84,
      "learning_rate": 8.108108108108109e-06,
      "loss": 0.1824,
      "step": 1280
    },
    {
      "epoch": 3.87,
      "learning_rate": 6.606606606606607e-06,
      "loss": 0.122,
      "step": 1290
    },
    {
      "epoch": 3.9,
      "learning_rate": 5.105105105105106e-06,
      "loss": 0.1079,
      "step": 1300
    },
    {
      "epoch": 3.9,
      "eval_accuracy": 0.9521785334750266,
      "eval_loss": 0.21390262246131897,
      "eval_runtime": 38.2092,
      "eval_samples_per_second": 24.628,
      "eval_steps_per_second": 3.088,
      "step": 1300
    },
    {
      "epoch": 3.93,
      "learning_rate": 3.603603603603604e-06,
      "loss": 0.1973,
      "step": 1310
    },
    {
      "epoch": 3.96,
      "learning_rate": 2.102102102102102e-06,
      "loss": 0.0976,
      "step": 1320
    },
    {
      "epoch": 3.99,
      "learning_rate": 6.006006006006006e-07,
      "loss": 0.1724,
      "step": 1330
    },
    {
      "epoch": 4.0,
      "step": 1332,
      "total_flos": 1.651775717862015e+18,
      "train_loss": 0.4871297548483084,
      "train_runtime": 1589.6821,
      "train_samples_per_second": 13.406,
      "train_steps_per_second": 0.838
    }
  ],
  "max_steps": 1332,
  "num_train_epochs": 4,
  "total_flos": 1.651775717862015e+18,
  "trial_name": null,
  "trial_params": null
}